My system had some bad RAM which I discovered after my docker containers started crashing. After running a ram test it turns out it was only 1 of the ram sticks that was bad, so I removed that bad one and running single channel at the moment (16GB). Ram test on the single slot passed (4 passes). However after booting back up, my system is not mounting my cache pool. It is telling me it is unmountable, I was following these documentations which tells me I need to do a scrub, however the scrub option is disabled because the drive is not mountable. (I think that doc needs updating). That leaves me with the only option of running a check in maintance mode with the --repair option, which has a huge warning not to do so. What can I do to get my system back up and running?

I added my diagnostic files.

On another note, I tried to see if removing one of the cache drives pool (the one I was seeing errors on in the logs) would allow me to mount at least one of them and copy some of the data that I have yet to backup. That didnt work, and now when I try to add it backup I get the warning that the drive will be formatted. I though unraid would be smart enough to know that the drive was the same as before and put them back in mirror mode without data lost, sigh.

I kept the array stopped for now. Hopefully you all can help me get back into mirror mode and be able to run a scrub, or somehow get them back to mountable state.

wanda-diagnostics-20250814-0158.zip

Oh, I do have a backup of the boot disk from when the cache was in mirror mode, if that helps in restoring the config without having to format the 2nd cache drive.

Scrub only runs on a mounted filesystem, and would not resolve this issue, you can try clearing the log, but there may be other issues:

btrfs rescue zero-log /dev/nvme0n1p1

Then restart the array and post new diags.

Thank you @JorgeB , shortly after I posted I found an old Ubuntu document that said to do that exact same thing. That got it back to be mountable. So I just made sure all my backups were up to date and I’m planning to just rebuild the cache when I return from work this evening.